Unreliable tests tell us nothing;Letter
This interpretation ignores two important facts. First the unreliability of the tests means at least a quarter of the children tested at key stage 2 get a "wrong" level. Therefore apparent lack of progress could be largely due to pupils being lucky at key stage 2 and unlucky at key stage 3.
More importantly, the tests at key stage 2 and 3 are not directly comparable, being based on completely different programmes of study. It is entirely possible for 11-year-olds to achieve Level 4, to make real progress in absolute terms, and then get to level 3 at 14, because they have not learnt enough of the new material in the Key Stage 3 curriculum.
The poor reliability and questionable validity of the tests simply do not allow us to draw meaningful conclusions about individuals. Even conclusions about cohorts need to be treated with caution. It may well be that pupil performance does dip in the first year of secondary school, but no amount of looking at national curriculum test data will establish this.
Professor of Education Assessment
Kings College London